Pediatric Extracorporeal Membrane Oxygenation Mortality Is Related to Extracorporeal Membrane Oxygenation Volume in US Hospitals

Abstract

Background: Our objective was to examine extracorporeal membrane oxygenation (ECMO) utilization and determine whether pediatric-specific and overall ECMO volumes are associated with mortality rates.

Methods: State Inpatient Databases from 17 states were queried for ECMO admissions during 2008-2014. Hospitals in which >90% of their ECMO patients were ≤18 y old were considered pediatric ECMO centers. Hospital overall ECMO volumes were calculated as the average annual number of admissions, of any age, and categorized as <6, 6-14, 15-30, and >30. Multivariable analyses were conducted to examine the impact of ECMO volume on pediatric in-hospital mortality.

Results: There were 4546 pediatric ECMO admissions across 84 hospitals. Most patients were neonates (59.9%), and the most common indication for ECMO was neonatal respiratory failure (20.1%). Approximately 35% of hospitals offering pediatric ECMO averaged <6 annual ECMO admissions. Centers with >30 annual ECMO admissions had significantly lower mortality than hospitals with lower ECMO volume. Among the high-volume centers, pediatric ECMO centers had significantly lower mortality rates than high-volume nonpediatric ECMO centers (17.4% versus 38.2%).

Conclusions: A high proportion of hospitals performing pediatric ECMO have a low number of annual ECMO admissions. Pediatric centers with high volume had the lowest risk-adjusted mortality rates for pediatric ECMO.

Keywords: ECMO; Mortality; Pediatric; Survival; Volume.